# AI video generation API for AI agents Scrollport gives AI agents 11 tools to generate videos, from ByteDance, MiniMax and xAI, through one connection. Prices are per 1k tokens or video second, depending on the tool, paid per use from one prepaid wallet with no subscription, provider accounts or API keys. Agents use them over MCP, the CLI or the HTTP API. Updated 6 October 2026 ## Choose an AI video model with Scrollport [Get started](https://scrollport.com/start) Use verified tools with pay-per-use pricing. Provider access is included. Tools that work in your own apps ask you to connect that account. Connect your first agent to receive $1 trial credit. ## Compare AI video generation API tools Prices are per 1k tokens or video second, depending on the tool, so compare what each tool returns and a typical run's cost before choosing. | Tool | Provider | What it returns | Price | | --- | --- | --- | --- | | [Seedance 1.5 Pro](https://scrollport.com/tools/providers/bytedance/seedance-1-5-pro) | ByteDance | Generate video from text prompts with duration, resolution and aspect-ratio controls. | $0.0034/1k tokens | | [Seedance 2.5 Image-to-Video](https://scrollport.com/tools/providers/bytedance/seedance-2-5-image) | ByteDance | Generate video with duration, resolution and native audio controls. | $0.0150/1k tokens | | [Seedance 2.5 Reference-to-Video](https://scrollport.com/tools/providers/bytedance/seedance-2-5-reference) | ByteDance | Generate video with duration, resolution and native audio controls. | $0.0150/1k tokens | | [Seedance 2.5 Text-to-Video](https://scrollport.com/tools/providers/bytedance/seedance-2-5-text) | ByteDance | Generate video with duration, resolution and native audio controls. | $0.0150/1k tokens | | [MiniMax H3 Reference-to-Video](https://scrollport.com/tools/providers/minimax/h3-reference-to-video) | MiniMax | Generate video guided by one to five reference images with duration, resolution and aspect-ratio controls. | $0.1120/video second | | [MiniMax H3 Text-to-Video](https://scrollport.com/tools/providers/minimax/h3-text-to-video) | MiniMax | Generate video from a text prompt with bounded duration, resolution and aspect-ratio controls. | $0.1120/video second | | [Grok Imagine Video 1.5 Image-to-Video](https://scrollport.com/tools/providers/xai/grok-imagine-video-1-5) | xAI | Generate video from an input image and text prompt with duration and resolution controls. | $0.1960/video second | | [Grok Imagine Video 1.5 Reference-to-Video](https://scrollport.com/tools/providers/xai/grok-imagine-video-1-5-reference-to-video) | xAI | Generate video from one to seven reference images and a tagged text prompt with duration and aspect-ratio controls. | $0.1960/video second | | [Grok Imagine Video 1.5 Text-to-Video](https://scrollport.com/tools/providers/xai/grok-imagine-video-1-5-text-to-video) | xAI | Generate video from a text prompt with duration, resolution and aspect-ratio controls. | $0.1960/video second | | [Seedance 2.5 Motion Reference](https://scrollport.com/tools/providers/bytedance/seedance-2-5-motion-reference) | ByteDance | Generate video guided by a source clip, with optional character images. | $0.2310/video second | | [Seedance 2.5 Portrait-to-Video](https://scrollport.com/tools/providers/bytedance/seedance-2-5-portrait) | ByteDance | Animate portrait references after automatic image review. | $0.2310/video second | ## 11 available tools Generate videos $0.0034/1k tokens ### Seedance 1.5 Pro Generate video from text prompts with duration, resolution and aspect-ratio controls. ByteDance Generate videos $0.0150/1k tokens ### Seedance 2.5 Image-to-Video Generate video with duration, resolution and native audio controls. ByteDance Generate videos $0.0150/1k tokens ### Seedance 2.5 Reference-to-Video Generate video with duration, resolution and native audio controls. ByteDance Generate videos $0.0150/1k tokens ### Seedance 2.5 Text-to-Video Generate video with duration, resolution and native audio controls. ByteDance Generate videos $0.1120/video second ### MiniMax H3 Reference-to-Video Generate video guided by one to five reference images with duration, resolution and aspect-ratio controls. MiniMax Generate videos $0.1120/video second ### MiniMax H3 Text-to-Video Generate video from a text prompt with bounded duration, resolution and aspect-ratio controls. MiniMax Generate videos $0.1960/video second ### Grok Imagine Video 1.5 Image-to-Video Generate video from an input image and text prompt with duration and resolution controls. xAI Generate videos $0.1960/video second ### Grok Imagine Video 1.5 Reference-to-Video Generate video from one to seven reference images and a tagged text prompt with duration and aspect-ratio controls. xAI Generate videos $0.1960/video second ### Grok Imagine Video 1.5 Text-to-Video Generate video from a text prompt with duration, resolution and aspect-ratio controls. xAI Generate videos $0.2310/video second ### Seedance 2.5 Motion Reference Generate video guided by a source clip, with optional character images. ByteDance Generate videos $0.2310/video second ### Seedance 2.5 Portrait-to-Video Animate portrait references after automatic image review. ByteDance ## Questions about AI video generation API ### How much do AI video generation API tools cost? Prices are per 1k tokens or video second, depending on the tool. You pay per use from one prepaid wallet with no subscription. Your agent sees the exact price before each run, and connecting your first agent adds $1 of trial credit. ### What is the best AI video generation API for an AI agent? It depends on what you need back and how much you will run it. Seedance 1.5 Pro (ByteDance, $0.0034/1k tokens): Generate video from text prompts with duration, resolution and aspect-ratio controls. Seedance 2.5 Image-to-Video (ByteDance, $0.0150/1k tokens): Generate video with duration, resolution and native audio controls. Seedance 2.5 Reference-to-Video (ByteDance, $0.0150/1k tokens): Generate video with duration, resolution and native audio controls. Your agent can inspect the closest match and see its exact price before it runs. ## Learn more about AI video generation Choose the production route from the assets you already have: a written shot, a source image or visual references. Your agent can then use the video tools that fit that starting point. **Should I use text-to-video, image-to-video or reference-to-video?** Use text-to-video when the shot starts from a written idea, image-to-video when one source frame should come alive, and reference-to-video when supplied visuals must guide identity, style or composition across the result. **Plan shots before generating a sequence** Break a concept into short shots with one subject action and one camera intention each. Define aspect ratio, duration and continuity requirements before generating expensive final footage. **Review continuity as well as visual quality** Check identity, wardrobe, objects, screen direction, lighting and transitions between shots. Give feedback on one observable problem at a time so the agent can revise the right part of the sequence. --- Canonical page: https://scrollport.com/tools/ai-video-generation-api